In June 1996, a 65-year-old man who lived in Marin County, California, became ill with blurred vision and dizziness. Eight days after the onset of symptoms, he visited his primary physician. A physical examination was remarkable only for nystagmus. Laboratory studies included leukocytes 8.2x10³, hematocrit 45.8%, and a chemistry panel that was normal except for phosphorus 4.5 mg/dL (normal range 2.7-4.4 mg/dL), cholesterol 265 mg/dL (normal range 125-200 mg/dL), and high-density lipoprotein cholesterol 67 mg/dL (normal range 35-60 mg/dL). A magnetic resonance image and arteriogram were normal. One month after the initial visit, the patient no longer complained of blurred vision or vertigo, and nystagmus had disappeared. Two years after the episode, he had no neurologic sequelae.

The patient lived in a suburban area of Marin County, approximately 1 km from a large brackish marsh bordering San Francisco Bay. He had traveled outside the United States the previous February, when he had visited Egypt and several Caribbean islands. He had not traveled outside California between this time and the onset of his illness in June, 4 months later.

An acute-phase serum specimen was sent to the Viral and Rickettsial Disease Laboratory, California Department of Health Services. Indirect immunofluorescence antibody tests were negative for St. Louis encephalitis and Western equine encephalomyelitis virus, as were serum samples taken 8 and 16 days after onset of illness. However, when this series of samples was tested by neutralization test 2 years later at the Arbovirus Research Unit of the University of California Center for Vector-Borne Disease Research as part of a retrospective study of arboviruses, the 8-day sample showed an antibody titer of 1:80 and the 16-day sample an antibody titer < 1:320 to California encephalitis (Table).

California encephalitis-related arboviruses included in the tests were snowshoe hare, La Crosse, Jamestown Canyon, Morro Bay, and Tahyna. Tahyna, an important California serogroup virus widely distributed in Europe and Asia, was included because of the patient's travel history. Northway, Main Drain, and Lokern viruses, members of the Bunyamwera serogroup occurring in California, were also included, as were Western equine encephalomyelitis virus and St. Louis encephalitis.

The sera were tested by plaque reduction-serum dilution neutralization with African green monkey kidney (Vero) cell cultures, according to the protocol of Campbell et al. (6). An increase in titer between the acute- and convalescent-phase samples was found only for California encephalitis and snowshoe hare virus. The titers for the convalescent-phase serum were 1:320 for California encephalitis and 1:180 for the closely related snowshoe hare virus (Table), which has never been reported in California. The patient had not traveled recently to alpine areas of the state, where snowshoe hare virus might be expected to occur.

Because the patient's travel occurred several months before his illness, exposure to mosquitoes near his home is the most likely route of infection. The most common mosquito species in salt marshes in Marin County are Aedes washinoi and Ae. squamiger (7). Bloodsucking adult females of both these species are usually present in late winter to early spring. Ae. dorsalis also occurs in this area, but later in the year. Culiseta inornata occurs frequently near salt marshes. California encephalitis has not been recovered from any of these species in California; most isolates have come from Ae. melanimon in the Central Valley (8).

Campbell et al. (5) found that most seropositive samples from humans, mostly from high elevations (>1,000 m) in California, were apparently infected with Jamestown Canyon virus. However, these investigators also reported that approximately 35% of samples from horses in low elevations (<1,000 m) in California showed evidence of prior infection with California encephalitis (9), but they did not test samples from the San Francisco Bay area. Fulhorst et al. (10) reported that 57% of the horses sampled in Marin County showed evidence of prior infection with Jamestown Canyon virus, but none with California encephalitis.

Awareness of arboviruses in the United States has been heightened as a result of the recent outbreak of human illnesses caused by West Nile virus. That outbreak was originally thought to be due to St. Louis encephalitis (11). The case we describe is a further reminder that clinicians should consider several causative agents when a patient has a central nervous system infection, especially if mosquito exposure has occurred.

Further studies are needed to assess the risk for human infection by California encephalitis in coastal California and the role of various mosquito species in transmission.